- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Link Copied
2 Replies
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Yes, you can link with the -L and -l specifications you mentioned. I agree that a few working examples would be useful. As a paying customer, you might make that suggestion on premier.intel.com.
There should be a short explanation of which libraries are in your version. Typically, for mkl, there are alternate versions, one requiring a p4 compatible CPU, one requiring p3 compatibility, and one "default." Then there is the one for lapack, and there is a libguide to support openmp in case a run-time kit is being installed separately from the Intel compiler, which already has that library. I hope I've represented it reasonably well.
With a given binary build, you can use ldd to find out which libraries it will require to run.
There should be a short explanation of which libraries are in your version. Typically, for mkl, there are alternate versions, one requiring a p4 compatible CPU, one requiring p3 compatibility, and one "default." Then there is the one for lapack, and there is a libguide to support openmp in case a run-time kit is being installed separately from the Intel compiler, which already has that library. I hope I've represented it reasonably well.
With a given binary build, you can use ldd to find out which libraries it will require to run.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Thanks.
I found very good info in /opt/intel/mkl60/doc/mkluse.htm
This info does not seem to be in the Intel Fortran Compiler User's Guide (which is what I was using) but is essential if one is to properly use the Intel math and lapack libraries.
I found very good info in /opt/intel/mkl60/doc/mkluse.htm
This info does not seem to be in the Intel Fortran Compiler User's Guide (which is what I was using) but is essential if one is to properly use the Intel math and lapack libraries.

Reply
Topic Options
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page